Question

我正在尝试移植RocAlphaGo来玩亚马逊游戏，并且在尝试实施有监督的政策培训师时会遇到问题。

from keras.models import Sequential, Model
from keras.layers.core import Activation, Flatten
from keras.layers import convolutional

defaults = {
            "board": 10,
            "filters_per_layer": 128,
            "layers": 12,
            "filter_width_1": 5
}
# copy defaults, but override with anything in kwargs
params = defaults
network = Sequential()
# create first layer
network.add(convolutional.Convolution2D(
      input_shape=(6, 10, 10),
      nb_filter=128,
      nb_row=5,
      nb_col=5,
      init='uniform',
      activation='relu',
      border_mode='same'))

# create all other layers
for i in range(2, 13):
     # use filter_width_K if it is there, otherwise use 3
     filter_key = "filter_width_%d" % i
     filter_width = params.get(filter_key, 3)

     # use filters_per_layer_K if it is there, otherwise use default value
     filter_count_key = "filters_per_layer_%d" % i
     filter_nb = params.get(filter_count_key, 128)

     network.add(convolutional.Convolution2D(
         nb_filter=filter_nb,
         nb_row=filter_width,
         nb_col=filter_width,
         init='uniform',
         activation='relu',
         border_mode='same'))

# the last layer maps each <filters_per_layer> feature to a number
network.add(convolutional.Convolution2D(
    nb_filter=1,
    nb_row=1,
    nb_col=1,
    init='uniform',
    border_mode='same'))
# reshape output to be board x board
network.add(Flatten())
# softmax makes it into a probability distribution
network.add(Activation('softmax'))

keras 1.2.0
python 2.7

给出以下例外：

ValueError：检查模型目标时出错：期望activation_1到有形状（无，60）但有阵形（10,100）

训练数据集是（10,6,10,10）阵列，10 x 6层，每层是10x10阵列（棋盘），为什么模型需要（无，60）？如果chagne input_shape=(6, 10, 10)到input_shape=(10, 10, 10)，将获得：

ValueError：检查模型输入时出错：预期 convolution2d_input_1有形状（无，10,10,10），但得到了数组形状（10,6,10,10）

所有代码均为here

Answer 1

正如Matias在评论中所说，如果你添加

network.summary()

您可能会注意到您的卷积应用于输入数据的前2个维度（即（6,10,10））。您的功能在第一维。默认情况下，当您使用tensorflow时，Keras将假定您的功能所在的维度是第三个，而不是第一个。因此，当你在（6,10,10）数组上应用128个过滤器时，输出将是（6,10,128），如果我理解的话，这不是你想要的。

由于这个原因，在最后一个卷积层的输出处，你得到一个（6,10,1）数组，它会像你期望的那样变平（，60）而不是（，100）。

有两种方法可以修复您的网络。您可以将输入数据更改为格式（10,10,6）。或者您使用data_format="channels_first"图层的Convolution2D()参数。

以下是第二个选项的代码：

from keras.models import Sequential, Model
from keras.layers.core import Activation, Flatten
from keras.layers import convolutional

defaults = {
            "board": 10,
            "filters_per_layer": 128,
            "layers": 12,
            "filter_width_1": 5
}
# copy defaults, but override with anything in kwargs
params = defaults
network = Sequential()
# create first layer
network.add(convolutional.Convolution2D(
      input_shape=(6, 10, 10),
      nb_filter=128,
      nb_row=5,
      nb_col=5,
      init='uniform',
      activation='relu',
      border_mode='same',
      data_format='channels_first'
      ))

# create all other layers
for i in range(2, 13):
     # use filter_width_K if it is there, otherwise use 3
     filter_key = "filter_width_%d" % i
     filter_width = params.get(filter_key, 3)

     # use filters_per_layer_K if it is there, otherwise use default value
     filter_count_key = "filters_per_layer_%d" % i
     filter_nb = params.get(filter_count_key, 128)

     network.add(convolutional.Convolution2D(
         nb_filter=filter_nb,
         nb_row=filter_width,
         nb_col=filter_width,
         init='uniform',
         activation='relu',
         border_mode='same',
         data_format='channels_first'))

# the last layer maps each <filters_per_layer> feature to a number
network.add(convolutional.Convolution2D(
    nb_filter=1,
    nb_row=1,
    nb_col=1,
    init='uniform',
    border_mode='same',
    data_format='channels_first'))
# reshape output to be board x board
network.add(Flatten())
# softmax makes it into a probability distribution
network.add(Activation('softmax'))
# display your network summary
network.summary()

修改

考虑到您的keras版本，您应该使用参数＆＃34; dim_ordering＆＃34;并将其设置为＆＃34; th＆＃34;。

我在keras documentation
中找到了这些信息
from keras.models import Sequential, Model from keras.layers.core import Activation, Flatten from keras.layers import convolutional defaults = { "board": 10, "filters_per_layer": 128, "layers": 12, "filter_width_1": 5 } # copy defaults, but override with anything in kwargs params = defaults network = Sequential() # create first layer network.add(convolutional.Convolution2D( input_shape=(6, 10, 10), nb_filter=128, nb_row=5, nb_col=5, init='uniform', activation='relu', border_mode='same', dim_ordering='th' )) # create all other layers for i in range(2, 13): # use filter_width_K if it is there, otherwise use 3 filter_key = "filter_width_%d" % i filter_width = params.get(filter_key, 3) # use filters_per_layer_K if it is there, otherwise use default value filter_count_key = "filters_per_layer_%d" % i filter_nb = params.get(filter_count_key, 128) network.add(convolutional.Convolution2D( nb_filter=filter_nb, nb_row=filter_width, nb_col=filter_width, init='uniform', activation='relu', border_mode='same', dim_ordering='th')) # the last layer maps each <filters_per_layer> feature to a number network.add(convolutional.Convolution2D( nb_filter=1, nb_row=1, nb_col=1, init='uniform', border_mode='same', dim_ordering='th')) # reshape output to be board x board network.add(Flatten()) # softmax makes it into a probability distribution network.add(Activation('softmax')) # display your network summary network.summary()

keras：ValueError：检查模型目标时出错：预期activation_1具有形状（无，60）但是有形状的数组（10,100）

1 个答案: